Eating Less but Gaining Weight? Why Under-Eating Backfires

No breakfast, because you're trying to be good. A small lunch. Chicken and vegetables for dinner. Then it's late, you're starving, and the night ends with a Magnum and a glass of wine.

Nutritionist Dr Federica Amati sees that day all the time. On Angela Foster's High Performance Health podcast, she described people who tell her they're putting on weight while hardly eating anything, and that's very often how their days go. Her advice surprises them: eat more. She says the look she gets is as if she's lost her mind.

Her bigger point is that food is only one of six things that hold up your health, and that they lean on each other. Here's that idea in 49 seconds.

Amati is a medical scientist and registered nutritionist, Head Nutritionist at ZOE, and she teaches at Imperial College London. Her new book is The Appetite Reset. This post is built on her ideas about calories, fullness and how we fuel ourselves, checked against the research. It covers why under-eating backfires, why calorie numbers are rougher than they look, and the six pillars in more detail.

Why under-eating all day backfires

From the inside, that evening feels like a willpower failure. You were good all day, then you blew it.

Amati sees it the other way round. The trouble starts in the morning. Eat very little and you don't give your body what it needs to feel full or, in her words, what it needs "to know that there isn't a famine coming". By the evening you're running on empty, and the quickest, most comforting food wins.

Two things are worth saying clearly. The Magnum isn't the villain here. One ice cream undoes nothing, and we've written about the guilt that can follow two Magnums. The trouble is the shape of the day, and the answer is more food, earlier, rather than more willpower at ten o'clock at night. It's the same reason CBT for binge eating starts with regular meals.

And if you're sure you really are eating less, Amati doesn't buy the idea that someone eats exactly the same for years and then, in a few months, everything changes. When she digs deeper, she says, it's rarely a like-for-like picture. That isn't an accusation. Part of it is that the numbers we use to judge "less" are rougher than they look, which we'll come to.

How your body knows it's full

Your gut lining is dotted with sensor cells. The best known are L cells, which release GLP-1 and PYY, two hormones that tell your brain you've had enough (Holst, 2007; Gribble and Reimann, 2016). Yes, that GLP-1: the hormone the weight-loss injections are designed to mimic.

These cells respond to what's in your food: protein, fat, carbohydrate and fibre. Fibre takes the long way round. Bacteria in your large intestine ferment it into short-chain fatty acids, and the sensor cells pick those up. When researchers at Imperial College London delivered one of those fatty acids, propionate, straight to people's colons, they released more PYY and GLP-1 and ate less afterwards (Chambers et al., 2015). Protein raises PYY too (Batterham et al., 2006).

So two days with the same calorie total can feel completely different. A day built on protein, plants and whole grains keeps sending "enough" signals. A day with very little food in it sends far fewer. Some foods also give you a lot of eating for very little energy. As Amati says, you'd need a huge amount of spinach to reach 200 calories, which is the whole idea behind big salads.

One of the clearest tests of food quality is a 2019 trial at the US National Institutes of Health. Twenty adults lived on a research ward for four weeks: two on ultra-processed food and two on unprocessed food, with the meals on offer matched for calories, sugar, fat, fibre and salt. They could eat as much or as little as they liked. On the ultra-processed diet they ate about 500 calories a day more and gained weight. On the unprocessed diet they lost weight (Hall et al., 2019). There's more on that trial in our ultra-processed food post.

How accurate are calorie labels?

Only roughly. Amati puts the error at about 20%, and that matches the rules. In the US, a label only counts as misbranded when the real calories are more than 20% above what it says (21 CFR 101.9). Guidance from the European Commission, written in 2012 while the UK was still in the EU, lets fat, carbohydrate and protein be 20% out in either direction on most foods (European Commission, 2012).

Real food bears this out. When Tufts University researchers tested reduced-calorie meals, restaurant dishes had 18% more calories than stated on average, and supermarket frozen meals 8% more (Urban et al., 2010).

You don't absorb everything on the label

How much energy you actually take in depends on the form the food is in. Almonds are the best-studied example. Chew a whole almond and most of its cells stay intact, with the fat sealed inside (Grundy et al., 2015). Grind it into a butter and the fat is free.

When USDA scientists fed people almonds in different forms and measured what their bodies absorbed, whole raw almonds gave about a third less energy than almond butter, gram for gram (Gebauer et al., 2016). An earlier USDA study found the standard calculation overstated the energy people got from almonds by 32% (Novotny et al., 2012).

What's on the rest of the plate matters as well. When USDA researchers added fibre to people's diets, they absorbed less of the fat and protein, so the same meals delivered less usable energy (Baer et al., 1997).

None of this makes calories meaningless. If you're a boxer making weight for a fight, a calorie deficit will get you there, and Amati says so. But for long-term health, she calls calories an imprecise tool. They tell you roughly how much energy a food holds, and nothing about how nourishing it is. If counting has never felt accurate to you, you can keep the Healthcount diary without any numbers at all.

Can you be thin and still unhealthy?

(Crisps are chips, if you're in the US.) Her point is that weight tells you very little on its own. Some people carry far more fat around their organs and in their liver than their size suggests. MRI scans of 477 adults at Imperial College London found exactly that, and the researchers named the pattern TOFI: thin on the outside, fat on the inside (Thomas et al., 2012).

Muscle is the other half of the story. When muscle is very low compared with fat, the risk of metabolic disease rises, a condition now formally defined as sarcopenic obesity (Donini et al., 2022). Muscle matters because it's where much of your blood sugar goes. In lab tests where insulin is held high, about 80% of the glucose cleared from the blood ends up in muscle (DeFronzo and Tripathy, 2009). In a US survey of more than 13,000 adults, people with more muscle for their size had better insulin sensitivity and less prediabetes and diabetes (Srikanthan and Karlamangla, 2011).

If you're on a GLP-1 medicine, this is why we keep coming back to protecting your muscle.

Why weight loss stalls when you eat too little

A calorie deficit works in the short term, and Amati is clear about that. Stay in one for a long time, though, and your body adapts. "Your body is extremely clever," she says. It lowers the energy it uses, makes you more tired so you move less, and weight loss slows to a stop.

This is one of the most consistent findings in obesity research. In a classic study at Rockefeller University in New York, people holding their weight 10% or more below where they started burned less energy each day than expected for their new size (Leibel et al., 1995). Most weight-loss attempts follow the same curve: early loss, then a plateau, then regain (Hall and Kahan, 2018).

She points out that the same plateau happens on GLP-1 medicines, and that early weight loss is partly water and partly lean mass, not only fat. We've written about why GLP-1 weight loss levels off and why your body defends its weight.

Eat more of the right foods

Her alternative is to eat as much of the right foods as you can. And "the right foods" is about adding, not cutting: more plants, some fermented foods, nuts and seeds, and whichever food groups are missing from your week.

Make breakfast the big one

Her simplest starting point is breakfast. Morning is when willpower feels strongest, which is exactly when people skip food. Amati suggests the opposite: make breakfast your most nutritious meal, with good protein, plenty of fibre, plant foods and some healthy fat, so you don't reach the evening starving.

The evidence here is worth being honest about. Eating breakfast doesn't make you lose weight by itself. A 2019 BMJ review of 13 trials found that people told to eat breakfast ate more over the day than people who skipped it (Sievert et al., 2019). But her point is about hunger later on, and there the evidence is more encouraging, if small. In a US trial of 20 young women who usually skipped breakfast, a protein-rich breakfast made them fuller through the day, raised PYY, lowered the hunger hormone ghrelin and cut evening snacking on high-fat foods. Their total calories didn't change (Leidy et al., 2013).

So think of breakfast as a way to make the evening easier, not a weight-loss trick. There's more on the protein side in eat the protein first.

The six pillars of health

Food is only one part of this, and this was the part of the conversation I found most useful. Lifestyle medicine, the branch of medicine that treats daily habits as part of care, groups health into the six pillars from the film at the top. Lifestyle medicine bodies in the UK and the US both use them, worded slightly differently.

Amati makes two points about them. First, they lean on each other. Her example: without good ways to handle stress, you might reach for a glass of wine to unwind. That hits your sleep. You feel rough the next day, so you skip seeing friends. One weak pillar puts weight on the others. That's the chain in the film.

Second, the basics come before the extras. Things like regular saunas are, in her words, the sprinkles on the cupcake. If you're working on those while only two of the six pillars are in place, it doesn't make sense.

Social connection is the pillar she thinks gets overlooked. She repeated something she'd seen online: you never see longevity influencers posting about hanging out with their friends or volunteering at their local community centre. Yet relationships are one of the strongest predictors of health we have. A review of 148 studies covering more than 300,000 people found that those with stronger social relationships had a 50% greater likelihood of survival over the studies' follow-up (Holt-Lunstad et al., 2010).

Together, the habits add up. In EPIC-Norfolk, a UK study of 20,244 adults followed for about 11 years, people with four healthy habits (not smoking, being physically active, drinking moderately and eating plenty of fruit and vegetables) had the same risk of dying as people 14 years younger who had none of them (Khaw et al., 2008). This study and the review of social relationships are both observational, so they show strong links rather than proof that the habits caused the difference.

Simple, but not easy

The pillars are free, in theory. In practice, Amati says, we live in an obesogenic environment: one that nudges us towards eating more and moving less. You can't walk down a street in London, she says, without passing shop after shop selling frappuccinos, muffins and fast food. In England, 64% of adults are living with overweight or obesity (Health Survey for England, 2022). In the US, 40% of adults are living with obesity (CDC, 2024). Figures that size say more about the world we live in than about anyone's willpower. There's more on this in how your environment affects your weight.

So her last point is about kindness. If building these habits takes real effort, that's because it does. Give yourself credit for trying, and give other people the same. And treat time as the one thing you can't buy more of: if a proper lunch break means blocking it out in your calendar, block it out.

What to take from this

If your evenings keep ending in a rescue snack, look at the morning rather than the night. Eat enough, earlier, and build meals around protein, plants and fibre. And look after the other pillars too, because sleep, stress and the people around you all shape what ends up on your plate.

A food diary is good at showing you the shape of a day, because it puts everything in order (here's why food diaries work). A notebook works. So does Healthcount, which lays the day out for you meal by meal.

If eating less has become a rule

If you recognised yourself in the restricting part of this, and cutting back has started to feel like a rule you can't break, you don't have to sort it out on your own. Talk to a doctor, or call the eating disorder helpline for your country. The one below follows the region set at the top of the page, and all four are in the sources.

We've also written about how CBT approaches food rules.

Questions people ask

This is general information, not medical advice. If you have a medical condition, are on a GLP-1 or are being treated for an eating disorder, your own clinicians' advice comes first.